Note: Descriptions are shown in the official language in which they were submitted.
. . . . ~.. . . .. . . . . ,... . .. . .
CA 02411628 2009-01-15
WO 02/06814 PCT/US01/22156
I
A SYSTEM AND METHOD OF D8T8thMaNCi POROSITY IN COMPOSITE
MATERIALS IISIltG IILTRASOIIND
TBCHNICAL FIlSI,D OF THE INVENTION
The present invention relates generally to an apparatus and method of
non-destructive evaluation of materials, especially composite materials, and
more particularly, to an apparatus and method for ultrasonically measuring the
porosity of a"composite material by accessing only one side of the composite
material.
BACKGROUND OF THF3 INVENTION
In recent years, the use of advanced composite structures has
experienced tremendous growth in the aerospace, automotive, and many other
commercial industries. While composite materials offer significant
improvements in performance, they require strict quality control procedures in
is the manufacturing processes. Specifically, non-destructive evaluation
("NDE")
methods are required to assess the structural integrity of composite
structures; for example, to detect inclusions, delaminations and porosities.
Conventional NDE methods, however,,are very slow, labor-intensive, and costly.
As a result, testing procedures adversely increase the manufacturing costs
associated with composite structures.
Various methods and systems have been proposed to assess the structural
integrity of composite structures. One method to generate and detect
ultrasound using lasers is disclosed in U.S. Patent 5,608,166, issued March 4,
1997, to Monchalin et al. (the '166 Patent"). The 1166 Patent discloses the
use of a first modulated, pulsed laser beam for generating ultrasound on a
work piece and a second pulsed laser beam for detecting the ultrasound. Phase
modulated light from the second laser beam is then demodulated to obtain a
signal representative of the ultrasonic motion at the surface of the work
piece. A disadvantage associated with this approach is that the first pulsed
laser beam must be modulated.
CA 02411628 2002-12-04
WO 02/06814 PCT/US01/22156
2
Fiber-reinforced polymer-matrix composite materials have been proved of
great interest in various fields for their high strength and low weight. In
the aircraft industry in particular, the use of these materials increases
steadily. However, polymer-matrix composites also must be inspected before
installation for the presence of defects such as delaminations, inclusions,
and porosity.
Porosity is the presence of voids created in composites during
manufacturing. Measurement of ultrasonic attenuation provides a measurement
of porosity level. Ultrasonic techniques were found to be the best technique
to assess porosity level in composites. Unfortunately, it is not always
possible to use a self-referencing technique exploiting two different echoes
in the signal, like two back-wall echoes. This is because in some cases, the
porosity is so high that barely one echo is observed. Porosity measurements
must rely on only one echo. Therefore, unlike delamination and inclusions,
the ultrasonic response to porosity is frequency and amplitude dependent.
This characteristic means that each ultrasonic system has a different response
to a given porosity level. Therefore, each ultrasonic system must be
calibrated relative to certain porosity levels.
Two different techniques are used to measure porosity: reflected through
transmission ultrasound (RTTU) and through transmission ultrasound (TTU).
These methods compare the amplitude of an ultrasonic wave that has propagated
through a sample with the amplitude of an ultrasonic wave that has propagated
through a reference sample. The ultrasonic echo used is the one reflected
from a steel or glass plate in the RTTU technique. FIGURE 1 depicts this
operation. Alternatively, in the TTU technique, ultrasonic echo is the echo
coming directly from the piezoelectric emitter. FIGURE 2 illustrates the TTU
Technique
Unfortunately, these techniques require an ultrasonic couplant, such as
water and access to both sides of each composite panels. Ultrasonic
refraction at the water/composite interface changes the ultrasonic wave
direction when the piezoelectric emitter is not normal to the sample surface.
Porosity must be evaluated at all points over the sample surface. The
ultrasonic system must be moved at several positions over the sample surface
to form a tight grid of ultrasonic measurements. The grid spacing must be
smaller than the size of a critical porosity defect. The normalcy requirement
makes the scanning of samples having curved surfaces expensive and slow.
Moreover, when the two sides of a composite part are not accessible, such as
for a closed box, porosity evaluations for such parts simply cannot be'made
using these techniques.
Therefore, a need has arisen for a method and system that permit making
porosity measurements and evaluations of composite materials, which corrects
the problems identified above, including the problems of needing water or
similar couplant or access to both sides of the composite material.
SUBSTITUTE SHEET (RULE 26)
CA 02411628 2002-12-04
WO 02/06814 PCT/US01/22156
3
Moreover, there is the need for a method and system that permit making
porosity measurements and evaluations of composite materials without the need
to modulate a pulsed laser beam or other similar problems such as those
existing with the 1166 Patent.
SUDMARY OF THE INVENTION
The present invention provides a method and system for ultrasonically
measuring the porosity in a composite material by accessing only one side of
the composite material that substantially eliminates or reduces disadvantages
and problems associated with previously developed ultrasonic systems and
methods.
The present invention provides a method and system to measure porosity
in composite material by ultrasonic techniques, such as laser ultrasound,
which method only requires access to one side of the composite material. The
method and system evaluate ultrasonic attenuation that correlates with
porosity, by comparing the amplitude of the sample material back-wall echo to
the amplitude of the back-wall echo of a reference sample. The method and
system further involves normalizing the signal amplitude according to a
reference echo (for laser-ultrasound, the surface echo) to eliminate amplitude
variations caused by phenomena not related to porosity.
According to another aspect of the invention, there is provided a method
and system for ultrasonically measuring the porosity in a sample composite
material by accessing only one side of the sample composite material. The
method includes the steps of measuring a sample ultrasonic signal from the
sample composite material, normalizing the sample ultrasonic signal relative
to the surface echo of the sample composite material, and isolating a sample
back-wall echo signal from the sample ultrasonic signal. A sample frequency
spectrum of said sample back-wall ultrasonic signal is then determined. Next,
the method and system include the steps of measuring a reference ultrasonic
signal from a reference composite material, normalizing said reference
ultrasonic signal relative to the surface echo of the reference composite
material and isolating a reference back-wall echo signal from said sample
ultrasonic signal. A reference frequency spectrum of said reference back-wall
ultrasonic signal is then determined. The invention further includes deriving
the ultrasonic attenuation of the sample ultrasonic signal as the ratio of the
sample frequency spectrum to the reference frequency spectrum over a
predetermined frequency range. Comparing the derived ultrasonic attenuation
to predetermined attenuation standards permits evaluating the porosity of the
sampled composite material.
A technical feature of the present invention is the use of the back-wall
echo amplitude of a composite material and comparing it to the back-wall echo
amplitude measured from a reference composite material. The present invention
normalizes all ultrasonic signals (including reference sample echo) relative
SUBSTITUTE SHEET (RULE 26)
CA 02411628 2002-12-04
WO 02/06814 PCT/US01/22156
4
to a reference echo (in the case of laser-ultrasound, the surface echo) to
eliminate amplitude variations caused by any effects not directly related to
porosity
Another technical feature of the present invention is the use of spatial
averaging of the ultrasonic traces to easily identify the position of the back
wall echo. The present invention includes calculating the full frequency
spectra of the sample back wall echo and of the reference back wall echo using
windowing techniques. The ratio of these two spectra provides the ultrasonic
attenuation as a function of frequency. The present invention uses the
ultrasonic attenuation at a given frequency to evaluate porosity relatively to
standards established previously for the material inspected, or to use the
frequency dependency of ultrasonic attenuation to correlate with porosity
level.
BRIEF DESCRIPTION OF THE DRAWINGS
For a more complete understanding of the present invention and the
advantages thereof, reference is now made to the following description taken
in conjunction with the accompanying drawings in which like reference numerals
indicate like features and wherein:
FIGUREs 1 and 2 show prior art methods of evaluating porosity in
composite materials;
FIGURE 3 illustrates the use of a generation laser beam and a detection
laser beam coaxial therewith;
FIGURE 4 is a block diagram showing the basic components of an apparatus
for performing ultrasonic laser testing;
FIGURE 5 shows an ultrasonic signal from a low porosity composite and an
ultrasonic signal obtained from a medium-to-severe porosity composite;
FIGURE 6 shows an example of laser-ultrasound signal normalization using
the surface echo;
FIGURE 7 presents the result of spatial averaging to identify the
arrival time of a back-wall echo in a medium-to-severe porosity composite.
FIGURE 8 presents the isolation of the back-wall echo for a composite
material according to the present invention;
FIGURE 9 presents the frequency spectra of a reference back-wall echo
and of the sample back-wall echo of FIGURE 8; and
FIGURE 10 shows the steps for performing ultrasonic attenuation
measurements for porosity measurements according to one embodiment of the
present invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
Preferred embodiments of the present invention are illustrated in
FIGUREs 3 through 10 of the drawings, like numerals being used for like and
corresponding parts of the various drawings.
SUBSTITUTE SHEET (RULE 26)
CA 02411628 2002-12-04
WO 02/06814 PCT/US01/22156
FIGURE 3 illustrates ultrasonic configuration 10; including incoming
laser beam 12 which represents a generation laser beam 14 and a coaxial
detection laser beam 16 upon a remote target 20. Generation laser beam 14
causes thermoelastic expansion in the target 20 in the form of ultrasonic
5 surface deformations, which deformations modulate, scatter and reflect
detection laser beam 16, represented by the phase-modulated light 18 directed
away from target 20.
FIGURE 4 illustrates in block diagram form the basic components of an
apparatus 21 for performing ultrasonic laser testing. Apparatus 21 comprises
a generation laser 22, a detection laser 24, an interferometer 230, an
optional optical processor 235, an optical scanner 240, collection optics 31,
systems controller 32, and data acquisition and processing apparatus 34.
Generation laser 22 and detection laser 24 generate a generation laser beam 14
and a detection laser beam 16, respectively, which are directed by optical
scanner'240 upon a target 20, which is typically a composite material. The
generation laser 22 produces a compressional ultrasonic wave in the material
normal to the surface of the target 20. The compressional ultrasonic wave is
the result of thermoelastic expansion of the composite material as it absorbs
generation laser beam 14.
The generation laser 22 must be of a frequency that is readily absorbed
into the surface of target 20 without causing ablation or breaking down the
target material, and it must be of the appropriate pulse duration to induce
ultrasonic surface deformations. For example, a transverse-excited
atmospheric ("TEA") COZ laser can be use d to produce a 10.6-micron wavelength
beam for a 100-nanosecond pulse. The power of the laser must be sufficient to
deliver, for example, a 0.25-joule pulse to the target, which may require a
100-watt laser operating at a 400 Hz pulse repetition rate. The generation
laser should be absorbed as heat into the target surface thereby causing
thermoelastic expansion without ablation.
The detection laser 24 must be of sufficient pulse duration to not
induce ultrasonic surface displacements. For example, a Nd:YAG laser can be
used. The power of this laser must be sufficient to deliver, for example, a
100 millijoule, 100-microsecond pulse, which may require a one-kilowatt laser.
U.S. Patent No. 6,122,060, entitled "METHOD AND APPARATUS FOR DETECTING
ULTRASONIC SURFACE DISPLACEMENTS USING POST-COLLECTION OPTICAL AMPLIFICATION,"
issued to Thomas E. Drake, Jr. on September 19, 2000 (the "Drake Patent") and
to which the present invention claims priority in part further describes the
above-illustrated system to include a detection laser to generate a first
pulsed laser beam to detect the ultrasonic surface displacements on a surface
of the target. Collection optics to collect phase modulated light from the
first pulsed laser beam either reflected or scattered by the target. An
optical amplifier which amplifies the phase modulated light collected by the
collection optics. An interferometer which processes the phase modulated
SUBSTITUTE SHEET (RULE 26)
. . 4. ... . . .. . .
CA 02411628 2009-01-15
WO 02/06814 PCT/US01/22156
6
light and generates at least one output eignal. A processor that processes
the at least one output signal to obtain data representative of the ultrasonic
surface displacements at the target. The present invention was developed for
the effort relating to Drake and may be used in such an environment.
The present invention measures the ultrasonic attenuation using only a
back-wall echo to evaluate porosity in composite parts. This prccedure is
useful for ultrasonic techniques requiring access to only one side of the
sample. The first step is to acquire the ultrasonic signal. FICUREs 5A and
5B presents ultrasonic signals obtained using laser-ultrasound in a composite
with low porosity, in FIGURE 5A, and in a composite with medium to severe
porosity, in FIGURE 5B.
Certain aspects of ultrasonic attenuation measurement are well-known
techniques for porosity evaluation. Ultrasonic evaluation of attenuation and,
using ultrasonic attenuation for solid-state material characterization, are
understood, as is the idea of using the ratio of frequency spectra to
calculate attenuation as a function of frequency is well known in the art.
Known techniques measure small ultrasonic attenuation variations. These
measurements are performed from only one side of the sample, but water
coupling is required and samples must have polished surfaces. In most cases of
using ultrasonic attenuation to evaluate porosity, either TTU technique or
RTTU technique are used. Both approaches use the ratio of two back wall
echoes to evaluate attenuation.
FIGUREs 5A and 5B show ultrasonic signals from a low porosity composite,
in FIGURE 5A, and ultrasonic signal obtained from a medium-to-severe porosity
composite in FIGURE SB. One difficulty with the use of a single back-wall
echo is that if the amplitude of the measured signal is affected by another
factor than porosity, porosity measurement can produce a false reading. The
ultrasonic signals need to be normalized in some manner. For this purpose, an
ultrasonic signal that is not affected by porosity can be used. In the case
of laser-ultrasound, this echo is called surface echo. This echo is produced
by the surface displacement created by thermal expansion. The surface echo
does not result from an ultrasonic wave. It is directly proportional to the
quantity of light collected by the detection system and to the efficiency of
the generation mechanism, assuming laser-ultrasound generation in an
homogeneous material and a flat-frequency response of the detection system.
Even if these conditions are not perfectly matched, normalization by the
surface provides in a first approximation a good correction for variations of
reflectivity and of generation and for variations of measurement incidence
angle. FIGURE 6 shows an example of laser-ultrasound signal normalization
usinq the surface echo.
CA 02411628 2002-12-04
WO 02/06814 PCT/US01/22156
7
The next step in the process of the present invention is to identify the
arrival time of the back-wall echo. In FIGURE 5A, the first back-wall echo is
easily identified. In the case of FIGURE 5B, on the other hand, the
identification of the back-wall echo is not easy. To identify the exact time
of the back-wall echo, the ultrasonic signals can be spatial averaged to
eliminate the spurious echoes caused by backscattering on porosity defects.
The spatially-averaged signal at a given position is obtained by averaging the
values of the signals at neighbor positions for each time. For example, the
spatial-averaged signal at position (50,50) is obtained by averaging for each
time values the ultrasonic signals between position (48,48) and (52,54) if a
kernel of 5x5 is used. The spatial averaging is used only to identify the
time of the back-wall echo. The back-wall echo arrival time of the spatial-
averaged signal is used on the non-averaged signal to window correctly the
back-wall echo in case sample thickness varies.
FIGUREs 7A and 7B present the result of spatially averaging to identify
the arrival time of a back-wall echo in a medium-to-severe porosity composite.
In particular, FIGURE 7A shows the ultrasonic signal without spatial
averaging, while FIGURE 7B depicts the ultrasonic signal at the same position
using spatial averaging with a 9x9 kernel. When the arrival time of the back-
wall echo has been found. This echo must be isolated from the remaining of
the signal. The echo is isolated using windowing techniques. Different types
of windowing techniques may be used, according to the given material and the
type of signal received
FIGURE 8 presents the isolation of the back-wall echo of FIGURE 7A. The
window position was obtained using the maximum of the analytic signal of
FIGURE 7B inside a time-window determined so that the surface echo does not
interfere. In particular, FIGURE 8 shows the effect of isolating back-wall
echo from the rest of the signal using classical windowing techniques. The
frequency spectrum of the back-wall echo is then calculated. This frequency
spectrum is then divided by the frequency spectrum of a reference signal. The
reference signal is obtained using the same experimental procedure and the
same signal-processing steps as the ones described previously for the signal
currently analyzed. The reference signal is obtained from a pristine sample
made of the same material with very low porosity and having the same thickness
than the sample currently tested.
FIGURE 9 presents the frequency spectra of a reference echo and of the
back-wall echo of FIGURE B. In FIGURE 9, the frequency spectra of a reference
echo is shown as solid line 40 and of the echo of FIGURE 6 is shown as dotted
line 42. The ratio of the two spectra gives the ultrasonic attenuation as a
function of frequency. To evaluate porosity, the ultrasonic attenuation at a
given frequency can be compared to pre-determined standards or the frequency
dependency of attenuation can be used.
SUBSTITUTE SHEET (RULE 26)
CA 02411628 2002-12-04
WO 02/06814 PCT/US01/22156
8
FIGURE 10 shows the steps required to perform ultrasonic attenuation
measurements for porosity measurements according to the present invention.
FIGURE 10 shows flow diagram 50 of the process of the present invention for
evaluating the porosity of a composite material. Referring to block 52, the
present invention involves the step of measuring the ultrasonic signal from a
composite material. This is done for both a reference composite material and a
sample composite material. Then, the present invention involves normalizing
each of these signal measurements, at block 54, with reference to amplitude of
the respective materials' surface echoes. Block 56 references the steps of
determining the back-wall echoes for both the normalized ultrasonic signals.
Then, at block 58, the process continues through the steps of calculating the
frequency spectrum of each of the normalized and measured signals. Block 60
accounts for the process step of dividing the frequency spectrum of the signal
by the frequency spectrum of the normalized reference back-wall echo. This
results in an ultrasonic attenuation measurement arising at a given frequency
or dependency of attenuation frequency.to evaluate the sample composite
material's porosity, as block 62 depicts.
The process of FIGURE 10 may be performed manually or automatically
through a computer and related ultrasonic sensing system, such as the system
described in FIGUREs 1 and 2 and in the Drake Patent. Accordingly, the
present invention encompasses a system fully capable of automatically
performing instructions and possessing the necessary data storage and
manipulation capabilities to carry out the described process.
Although the present invention has been described in detail, it should
be understood that various changes, substitutions and alterations can be made
hereto without departing from the spirit and scope of the invention as defined
by the appended claims.
SUBSTITUTE SHEET (RULE 26)
